In both humans and mice, astrocytes form the immediate cellular environment of the ganglion cell axons in the optic nerve head. They occupy up to 50% of the total tissue volume at that location. When the axons of retinal ganglion cells are damaged, they become reactive. In the proposed research we investigate the signaling between optic axons and the astrocyte that triggers this response. The first question is the nature and timing of the changes in gene expression that follow damage to the optic axons. Our preliminary results indicate that very different sets of glial genes are up or down regulated at different times following axonal injury. The experiments of Specific Aim 1 seek to confirm this finding, using an improved model of axonal damage. Specific Aim 2 uses the results of our existing and future expression profiling to identify candidates for the signaling systems that communicate from axons to astrocytes. The candidate signals will be overexpressed, using viral vectors, in retinal ganglion cells. The prediction is that overexpression of an actual signal will lead to the changes in astrocyte gene expression and morphology characteristic of astrocyte reactivity. Specific Aim 3 investigates the hypothesis that the initial response of the astrocytes to axonal damage is a protective one, i.e. that retinal ganglion cells threatened with injury send a distress signal to the ensheathing astrocytes, which then initiate a neuroprotective response. We will test this hypothesis by stressing the optic nerve in transgenic mice carrying mutations that compromise astrocyte communication and motility. The prediction is that the compromise of astrocyte function will worsen the damage to the axons resulting from this stress.